Multivariate outlier detection based on a robust Mahalanobis distance with shrinkage estimators

DOI10.1007/S00362-019-01148-1zbMATH Open1477.62133arXiv1904.02596OpenAlexW3098891640WikidataQ126805938 ScholiaQ126805938MaRDI QIDQ2065296FDOQ2065296

Authors: Elisa Cabana, Henry Laniado, Rosa E. Lillo 

Publication date: 7 January 2022

Published in: Statistical Papers (Search for Journal in Brave)

Abstract: A collection of robust Mahalanobis distances for multivariate outlier detection is proposed, based on the notion of shrinkage. Robust intensity and scaling factors are optimally estimated to define the shrinkage. Some properties are investigated, such as affine equivariance and breakdown value. The performance of the proposal is illustrated through the comparison to other techniques from the literature, in a simulation study and with a real dataset. The behavior when the underlying distribution is heavy-tailed or skewed, shows the appropriateness of the method when we deviate from the common assumption of normality. The resulting high correct detection rates and low false detection rates in the vast majority of cases, as well as the significantly smaller computation time shows the advantages of our proposal.

Full work available at URL: https://arxiv.org/abs/1904.02596

Recommendations

• Mahalanobis distance and its application for detecting multivariate outliers
• Detection of outliers in multivariate data: a method based on clustering and robust estimators
• Multidimensional outlier detection and robust estimation using \(S_n\) covariance
• Multivariate outliers and decompositions of mahalanobis distance
• On robustness of the test for detection of multivariate outliers
• A New Approach for Detecting Multivariate Outliers
• scientific article; zbMATH DE number 4020243

zbMATH Keywords

comedian matrixmultivariate \(L_1\)-medianmultivariate distancerobust location and covariance matrix estimation

Mathematics Subject Classification ID

Characterization and structure theory for multivariate probability distributions; copulas (62H05) Estimation in multivariate analysis (62H12) Ridge regression; shrinkage estimators (Lasso) (62J07) Robustness and adaptive procedures (parametric inference) (62F35)

Cites Work

• Title not available (Why is that?)
• The Masking Breakdown Point of Multivariate Outlier Identification Rules
• Propagation of outliers in multivariate data
• A well-conditioned estimator for large-dimensional covariance matrices
• Title not available (Why is that?)
• Robust Estimation of Dispersion Matrices and Principal Components
• The multivariate L ₁ -median and associated data depth
• Title not available (Why is that?)
• Title not available (Why is that?)
• Breakdown points of affine equivariant estimators of multivariate location and covariance matrices
• High-breakdown robust multivariate methods
• On a notion of data depth based on random simplices
• A two-sample test for high-dimensional data with applications to gene-set testing
• Estimation with quadratic loss.
• Title not available (Why is that?)
• The statistical analysis of fMRI data
• Robustness and Complex Data Structures
• From Depth to Local Depth: A Focus on Centrality
• Local depth
• Multivariate nonparametric methods with R. An approach based on spatial signs and ranks.
• Robust Shrinkage Estimation of High-Dimensional Covariance Matrices
• Parallel Coordinates
• The statistical analysis of functional MRI data
• Large dimensional analysis and optimization of robust shrinkage covariance matrix estimators
• Title not available (Why is that?)
• Robust and efficient estimation of multivariate scatter and location
• Robust estimation of precision matrices under cellwise contamination
• Estimating the asymptotic dispersion of the \(L_ 1\) median
• Title not available (Why is that?)
• On the Distribution of the Sample Median
• Limit theorems for the median deviation
• On MAD and comedians
• On the dispersion of multivariate median
• Shrinkage for covariance estimation: asymptotics, confidence intervals, bounds and applications in sensor monitoring and finance
• Outlier detection for high dimensional data using the comedian approach

Cited In (14)

• On robust Mahalanobis distance issued from minimum vector variance
• Robust distances for outlier-free goodness-of-fit testing
• Multivariate outliers and decompositions of mahalanobis distance
• Three estimators of the Mahalanobis distance in high-dimensional data
• On masking and swamping robustness of leading nonparametric outlier identifiers for multivariate data
• Outlier detection for compositional data using robust methods
• A cluster-based outlier detection scheme for multivariate data
• Outlier detection with Mahalanobis square distance: incorporating small sample correction factor
• Mahalanobis distance and its application for detecting multivariate outliers
• A further study comparing forward search multivariate outlier methods including ATLA with an application to clustering
• Multidimensional outlier detection and robust estimation using \(S_n\) covariance
• Enhancing multivariate control charts for individual observations using ROC estimates
• Robust modified classical spherical tests in the presence of outliers
• Geographically weighted Comedian method for spatial outlier detection

Uses Software

• AS 78
• MNM
• LIBRA

This page was built for publication: Multivariate outlier detection based on a robust Mahalanobis distance with shrinkage estimators

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2065296)